• Wind and Structures
• /
• 제3권2호
• /
• pp.117-132
• /
• 2000

Under special conditions of turbulent wind, suspension and cable-stayed bridges could reach instability conditions. In various instances the bridge deck, as like a bluff body, could exhibit single-degree torsional instability. In the present study the turbulent component of flow has been considered as a solution of a differential stochastic linear equation. The input process is represented by a Gaussian zero-mean white noise. In this paper the analytical solution of the dynamic response of the bridge has been determined. The solution has been obtained with a technique of closing on the order of the moments.

• 대한전기학회논문지
• /
• 제39권7호
• /
• pp.729-739
• /
• 1990

This paper deals with the fault detection problem in uncertain linear/non-linear systems having both undermodelling and noise. A robust fault detection method is presented which accounts for the effects of noise, model mismatch and nonlinearities. The basic idea is to embed the unmodelled dynamics in a stochastic process and to use the nominal model with a predetermined fixed denominator. This allows the input /output relationship to be represented as a linear function of the system parameters and also facilitate the quatification of the effect of noise, model mismatch and linearization errors on parameter estimation by the Bayesian method. Comparisons are made via simulations with traditional fault detection methods which do not account for model mismatch or linearization errors. The new method suggested in this paper is shown to have a marked improvement over traditional methods on a number of simulations, which is a consequence of the fact that the new method explicitly for the effects of undermodelling and linearization errors.

• 대한수학회논문집
• /
• 제21권3호
• /
• pp.543-558
• /
• 2006

We consider a type of large deviation Principle(LDP) using Freidlin-Wentzell exponential estimates for the solutions to perturbed stochastic differential equations(SDEs) driven by Martingale measure(Gaussian noise). We are using exponential tail estimates and exit probability of a diffusion process. Referring to Freidlin-Wentzell inequality, we want to show another approach to get LDP for the solutions to SDEs.

• 대한전기학회논문지
• /
• 제35권11호
• /
• pp.503-510
• /
• 1986

A new algorithm based on the concept of prediction error minimization is suggested to estimate the time varying delay in discrete processes. In spite of the existence of the stochastic noise, this algorithm can estimate time varying delay accurately. Computation time of this algorithm is far less than that of the previous extended parameter methods. With the use of this algorithm, generalized minimum variance control shows good control behavior in simulations.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.469-474
• /
• 2002

In recent years, modularization of engine parts has increased the application of plastic products in air intake systems. Plastic intake manifolds provide many advantages including reduced weight, contracted cost, and lower intake air temperatures. These manifolds, however, have some weakness when compared with customary aluminium intake manifolds, in that they have low sound transmission loss because of their lower material density. This low transmission loss of plastic intake manifolds causes several problems related to flow noise, especially when the throttle is opened quickly. The physical processes, responsible for this flow noise, include turbulent fluid motion and relative motion of the throttle to the airflow. The former is generated by high-speed airflow in the splits between the throttle valve and the inner-surface of the throttle body and surge-tank, which can be categorized into the quadrupole source. The latter induces the unsteady force on the flow, which can be classified into the dipole source. In this paper, the mechanism of noise generation from the turbulence is only investigated as a preliminary study. Stochastic noise source synthesis method is adopted for the analysis of turbulence-induced, i.e. quadrupole noise by throttle at quick opening state. The method consists of three procedures. The first step corresponds to the preliminary time-averaged Navier-Stokes computation with a $k-\varepsilon$ turbulence model providing mean flow field characteristics. The second step is the synthesis of time-dependent turbulent velocity field associated with quadrupole noise sources. The final step is devoted to the determination of acoustic source terms associated with turbulent velocity. For the first step, we used market available analysis tools such as STAR-CD, the trade names of fluid analysis tools available on the market. The steady state flows at three open angle of throttle valve, i.e. 20, 35 and 60 degree, are numerically analyzed. Then, time-dependent turbulent velocity fields are produced by using the stochastic model and the flow analysis results. Using this turbulent velocity field, the turbulence-originated noise sources, i.e. the self-noise and shear-noise sources are synthesized. Based on these numerical results, it is found that the origin of the turbulent flow and noise might be attributed to the process of formulation and the interaction of two vortex lines formed in the downstream of the throttle valve. These vortex lines are produced by the non-uniform splits between the throttle valve and inner cylinder surface. Based on the analysis, we present the low-noise design of the inner geometry of throttle body.

• Transactions on Control, Automation and Systems Engineering
• /
• 제2권1호
• /
• pp.40-46
• /
• 2000

In this paper, a transient identification based on a Hidden Markov Model (HMM) has been suggested and evaluated experimentally for the classification of transients in the dynamic process. The transient can be identified by its unique time dependent patterns related to the principal variables. The HMM, a double stochastic process, can be applied to transient identification which is a spatial and temporal classification problem under a statistical pattern recognition framework. The HMM is created for each transient from a set of training data by the maximum-likelihood estimation method. The transient identification is determined by calculating which model has the highest probability for the given test data. Several experimental tests have been performed with normalization methods, clustering algorithms, and a number of states in HMM. Several experimental tests have been performed including superimposing random noise, adding systematic error, and untrained transients. The proposed real-time transient identification system has many advantages, however, there are still a lot of problems that should be solved to apply to a real dynamic process. Further efforts are being made to improve the system performance and robustness to demonstrate reliability and accuracy to the required level.

• 한국정보기술응용학회:학술대회논문집
• /
• 한국정보기술응용학회 2002년도 추계공동학술대회 정보환경 변화에 따른 신정보기술 패러다임
• /
• pp.461-467
• /
• 2002

뇌에서 얻은 시계열 신호들은 대부분 불규칙하고 복잡한 파형을 가지고 있으며, 1980년대 중반까지만 해도 이러한 신호들은 확률론 과정(stochastic process)으로 발생된 '소음'(noise)으로 여겨졌다. 하지만 최근 들어 뇌에 관한 연구가 활발히 진행되었고 EEG를 이용한 생체신호의 비선형분석에 관한 연구가 진행되면서 뇌에서 발생되는 신호는 의미 있는 신호로 분석되어지고 있다. 이에 비선형 분석방법인 상관차원을 이용하여 라벤더 향기 전과 향기 후의 뇌파의 변화를 분석하고, 주파수 대역별로 delta파, theta파, alpha파, beta파로 나누어서 라벤더 향이 뇌에 미치는 영향을 분석한다. 즉, 뇌에서 발생되는 신호의 특징을 찾기 위해 다른 향보다 강하게 반응하는 라벤더 향을 후각자극으로 제시하여 EEG를 측정한 후, 16채널에 대한 상관차원을 구하고 라벤더 향이 뇌에 미치는 영향을 분석한다.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2002년도 추계공동학술대회
• /
• pp.461-467
• /
• 2002

뇌에서 얻은 시계열 신호들은 대부분 불규칙하고 복잡한 파형을 가지고 있으며, 1980년대 중반까지만 해도 이러한 신호들은 확률론 과정(stochastic process)으로 발생된 '소음'(noise)으로 여겨졌다. 하지만 최근 들어 뇌에 관한 연구가 활발히 진행되었고 EEG를 이용한 생체신호의 비선형 분석에 관한 연구가 진행되면서 뇌에서 발생되는 신호는 의미 있는 신호로 분석되어지고 있다. 이에 비선형 분석방법인 상관차원을 이용하여 라벤더 향기 전과 향기 후의 뇌파의 변화를 분석하고, 주파수 대역별로 delta파, theta파, alpha파, beta파로 나누어서 라벤더 향이 뇌에 미치는 영향을 분석한다. 즉, 뇌에서 발생되는 신호의 특징을 찾기 위해 다른 향보다 강하게 반응하는 라벤더 향을 후각 자극으로 제시하여 EEG를 측정한 후, 16채널에 대한 상관차원을 구하고 라벤더 향이 뇌에 미치는 영향을 분석한다.

• 방송공학회논문지
• /
• 제5권2호
• /
• pp.281-289
• /
• 2000

영상을 획득하는 과정에 있어서 영상 획득 장치 또는 피사체의 흔들림은 영상에 손상을 가져온다. 이러한 손상을 움직임 열화(motion blur)라고 부르며, 영상의 선명도를 떨어뜨리는 주된 원인이 된다. 최근 연구에서 밝힌 극점자취 방법을 통해 주어진 열화영상에서 열화의 PSF(Point Spread Function) 특성을 구하는데 사용되는 중요한 파라메터를 추출 할 수 있다. 이러한 극점 자취방법으로, 노이즈에 의한 열화에 관계없이 적은 연산량으로 움직임 열화의 방향을 추출할 수 있다. 본 논문에서는 통계적 극점 자취 접근법을 새롭게 제안한다. 저주파 영역에서 움직임 열화방향의 추정오차를 줄이기 위해 ML(Maximum Likelihood)분류를 통해 오차를 유발하는 극점을 선택하여 가중치를 적용, 그 영향을 최소화한다. 선형 예측법을 사용하여, 불규칙적 자료가 극점으로 선택되는 것을 방지한다 제안된 MALM(Moving average least mean)방법은 두번째로 큰 극점의 검출을 위해 움직임의 정도를 판별하는데 사용된다. MALM방법은 자체적으로 노이즈 제거 과정을 내포하고 있으므로 노이즈가 많은 환경에서도 파라메터를 추출할 수가 있다. 실험에서 우리는 제안된 방법을 통해 얻어진 정보를 사용하여, 열화 된 이미지를 효율적으로 복구해 낼 수 있었다.

• 한국소음진동공학회논문집
• /
• 제19권9호
• /
• pp.884-891
• /
• 2009

This study investigates some of swarm intelligence algorithms to tackle a traditional damage detection problem having stiffness degradation or damage in mechanical structures. Particle swarm(PSO) and ant colony optimization(ACO) methods have been exploited for localizing and estimating the location and extent damages in a structure. Both PSO and ACO are population-based, stochastic algorithms that have been developed from the underlying concept of swarm intelligence and search heuristic. A finite element (FE) model updating is implemented to minimize the difference in a set of natural frequencies between measured and baseline vibration data. Stiffness loss of certain elements is considered to simulate structural damages in the FE model. It is numerically shown that PSO and ACO algorithms successfully completed the optimization process of model updating in locating unknown damages in a truss structure.